INDEX

Explanations

references to books and educational publications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

author of books and papers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MigrationBuilder

-0.61

AutoScaleMode

-0.59

protoimpl

-0.57

CloseOperation

-0.57

 disambiguazione

-0.54

 Numerade

-0.53

Chham

-0.51

Derbyniad

-0.51

şiv

-0.50

RTEE

-0.50

POSITIVE LOGITS

 cible

0.40

 target

0.40

 textbooks

0.38

 ditu

0.35

 textbook

0.35

 zawier

0.34

 Marshal

0.34

 Actor

0.33

target

0.33

 procedural

0.33

Activations Density 0.022%

references to books and educational publications

author of books and papers

No Comments

No Known Activations